UnDeepVO: Monocular Visual Odometry Through Unsupervised Deep Learning

Li, Ruihao; Wang, Sen; Li, Zhiqiang; Gu, Dongbing

doi:10.1109/icra.2018.8461251

Cited by 476 publications

(379 citation statements)

References 18 publications

Supporting

Mentioning

373

Contrasting

Unclassified

Order By: Relevance

“…The network architecture consists of CNN and LSTM, and the LSTM was proved to help reduce noises by incorporating the temporal information from past sequences. Li et al [14] presented UnDeepVO to estimate the 6-DOF poses of a monocular camera and the depth of its view. Stereo image pairs were harnessed for recovering absolute scale and loss function was defined on spatial and temporal dense information.…”

Section: A Deep Learning For Visual Odometrymentioning

confidence: 99%

DeepPCO: End-to-End Point Cloud Odometry through Deep Parallel Neural Network

Wang

Saputra

Zhao

et al. 2019

2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

Odometry is of key importance for localization in the absence of a map. There is considerable work in the area of visual odometry (VO), and recent advances in deep learning have brought novel approaches to VO, which directly learn salient features from raw images. These learning-based approaches have led to more accurate and robust VO systems. However, they have not been well applied to point cloud data yet. In this work, we investigate how to exploit deep learning to estimate point cloud odometry (PCO), which may serve as a critical component in point cloud-based downstream tasks or learning-based systems. Specifically, we propose a novel end-to-end deep parallel neural network called DeepPCO, which can estimate the 6-DOF poses using consecutive point clouds. It consists of two parallel sub-networks to estimate 3-D translation and orientation respectively rather than a single neural network. We validate our approach on KITTI Visual Odometry/SLAM benchmark dataset with different baselines. Experiments demonstrate that the proposed approach achieves good performance in terms of pose accuracy.

show abstract

Section: A Deep Learning For Visual Odometrymentioning

confidence: 99%

DeepPCO: End-to-End Point Cloud Odometry through Deep Parallel Neural Network

Wang

Saputra

Zhao

et al. 2019

2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

show abstract

“…[2] utilizes synthetic data and style transfer [55] to collect more diverse training data. [26] uses stereo image pairs to recover the scale.…”

Section: Related Workmentioning

confidence: 99%

Unsupervised High-Resolution Depth Learning From Videos With Dual Networks

Zhou

Wang

Qin

et al. 2019

2019 IEEE/CVF International Conference on Computer Vision (ICCV)

View full text Add to dashboard Cite

Unsupervised depth learning takes the appearance difference between a target view and a view synthesized from its adjacent frame as supervisory signal. Since the supervisory signal only comes from images themselves, the resolution of training data significantly impacts the performance. High-resolution images contain more fine-grained details and provide more accurate supervisory signal. However, due to the limitation of memory and computation power, the original images are typically down-sampled during training, which suffers heavy loss of details and disparity accuracy. In order to fully explore the information contained in high-resolution data, we propose a simple yet effective dual networks architecture, which can directly take highresolution images as input and generate high-resolution and high-accuracy depth map efficiently. We also propose a Self-assembled Attention (SA-Attention) module to handle low-texture region. The evaluation on the benchmark KITTI and Make3D datasets demonstrates that our method achieves state-of-the-art results in the monocular depth estimation task.

show abstract

“…In contrast to the previous approaches that utilize the ground truth depth as supervision, some works [6], [19] utilize the task of view synthesis as supervision for single-view depth estimation. This view synthesis approach was extended to unsupervised end-to-end learning of both depth and egomotion estimation [12], [20], [21], [22]. These methods simultaneously train a depth and a pose network on temporally contiguous frames with a photometric loss between the target and a synthesized nearby view.…”

Section: A Depth From a Single Imagementioning

confidence: 99%

Learning Residual Flow as Dynamic Motion from Stereo Videos

Lee

Lin

et al. 2019

2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

We present a method for decomposing the 3D scene flow observed from a moving stereo rig into stationary scene elements and dynamic object motion. Our unsupervised learning framework jointly reasons about the camera motion, optical flow, and 3D motion of moving objects. Three cooperating networks predict stereo matching, camera motion, and residual flow, which represents the flow component due to object motion and not from camera motion. Based on rigid projective geometry, the estimated stereo depth is used to guide the camera motion estimation, and the depth and camera motion are used to guide the residual flow estimation. We also explicitly estimate the 3D scene flow of dynamic objects based on the residual flow and scene depth. Experiments on the KITTI dataset demonstrate the effectiveness of our approach and show that our method outperforms other state-of-the-art algorithms on the optical flow and visual odometry tasks. † Part of this work was done during an internship at Microsoft Research Asia 1 S. Lee, S. Im and I. S. Kweon are with the Robotics and Computer Vision Laboratory, KAIST, Daejeon, 34141, Republic of Korea {snapillar, dlarl8927, iskweon77}@kaist.ac.kr 2 S. Lin is with the

show abstract

UnDeepVO: Monocular Visual Odometry Through Unsupervised Deep Learning

Cited by 476 publications

References 18 publications

DeepPCO: End-to-End Point Cloud Odometry through Deep Parallel Neural Network

DeepPCO: End-to-End Point Cloud Odometry through Deep Parallel Neural Network

Unsupervised High-Resolution Depth Learning From Videos With Dual Networks

Learning Residual Flow as Dynamic Motion from Stereo Videos

Contact Info

Product

Resources

About